The sediment transport, involving the movement of the bedload and suspended sediment in the basins, is a critical environmental concern that worsens water scarcity and leads to degradation of land and its ecosystems. Machine learning (ML) algorithms have emerged as powerful tools for predicting sediment yield. However, their use by decision-makers can be attributed to concerns regarding their consistency with the involved physical processes. In light of this issue, this study aims to develop a physics-informed ML approach for predicting sediment yield. To achieve this objective, Gaussian, Center, Regular, and Direct Copulas were employed to generate virtual combinations of physical of the sub-basins and hydrological datasets. These datasets were then utilized to train deep neural network (DNN), conventional neural network (CNN), Extra Tree, and XGBoost (XGB) models. The performance of these models was compared with the modified universal soil loss equation (MUSLE), which serves as a process-based model. The results demonstrated that the ML models outperformed the MUSLE model, exhibiting improvements in Nash–Sutcliffe efficiency (NSE) of approximately 10%, 18%, 32%, and 41% for the DNN, CNN, Extra Tree, and XGB models, respectively. Furthermore, through Sobol sensitivity and Shapley additive explanation–based interpretability analyses, it was revealed that the Extra Tree model displayed greater consistency with the physical processes underlying sediment transport as modeled by MUSLE. The proposed framework provides new insights into enhancing the accuracy and applicability of ML models in forecasting sediment yield while maintaining consistency with natural processes. Consequently, it can prove valuable in simulating process-related strategies aimed at mitigating sediment transport at watershed scales, such as the implementation of best management practices.

The airway macrophages carbon loading (AMCL) has been suggested to be a biomarker of the long-term exposure to air pollution; however, to date no study has characterized AMCL for the pregnancy period. Therefore, this study aimed to assess the determinants of AMCL during pregnancy in Iran, a middle-income country. This study was based on a sample of 234 pregnant women with term and normal vaginal delivery who were residing in Sabzevar, Iran (2019). We characterized 35 potential determinants of personal exposure to air pollution for each participant, including six personal, nine indoor, and 20 home-outdoor factors. We applied Deletion/Substitution/Addition algorithm to identify the most relevant determinants that could predict AMCL levels. The median (IQR) of AMCL level was 0.12 (0.30) μm² with a successful sputum induction in 82.9% (194) of participants. Ambient residential PM₂.₅ levels were positively associated with higher AMCL levels. On the other hand, increased residential distance to the traffic lights, squares and ring-roads, the duration of opening window per day, and opening window during cooking were inversely associated with AMCL levels. Our findings provide novel insights on the different personal, indoor, and outdoor determinants of personal exposure to air pollution during pregnancy in a middle-income country.

Most amyotrophic lateral sclerosis (ALS) cases are sporadic (∼90%) and environmental exposures are implicated in their etiology. Large industrial facilities are permitted the airborne release of certain chemicals with hazardous properties and report the amounts to the US Environmental Protection Agency (EPA) as part of its Toxics Release Inventory (TRI) monitoring program. The objective of this project was to identify industrial chemicals released into the air that may be associated with ALS etiology. We geospatially estimated residential exposure to contaminants using a de-identified medical claims database, the SYMPHONY Integrated Dataverse®, with ∼26,000 nationally distributed ALS patients, and non-ALS controls matched for age and gender. We mapped TRI data on industrial releases of 523 airborne contaminants to estimate local residential exposure and used a dynamic categorization algorithm to solve the problem of zero-inflation in the dataset. In an independent validation study, we used residential histories to estimate exposure in each year prior to diagnosis. Air releases with positive associations in both the SYMPHONY analysis and the spatio-temporal validation study included styrene (false discovery rate (FDR) 5.4e-5), chromium (FDR 2.4e-4), nickel (FDR 1.6e-3), and dichloromethane (FDR 4.8e-4). Using a large de-identified healthcare claims dataset, we identified geospatial environmental contaminants associated with ALS. The analytic pipeline used may be applied to other diseases and identify novel targets for exposure mitigation. Our results support the future evaluation of these environmental chemicals as potential etiologic contributors to sporadic ALS risk.

One of the fundamental sustainable development goals has been recognized as having access to clean water for drinking purposes. In the Anthropocene era, rapid urbanization put further stress on water resources, and associated groundwater contamination expanded into a significant global environmental issue. Natural arsenic and related water pollution have already caused a burden issue on groundwater vulnerability and corresponding health hazard in and around the Ganges delta. A field based hydrogeochemical analysis has been carried out in the elevated arsenic prone areas of moribund Ganges delta, West Bengal, a part of western Ganga- Brahmaputra delta (GBD). New data driven heuristic algorithms are rarely used in groundwater vulnerability studies, specifically not yet used in the elevated arsenic prone areas of Ganges delta, India. Therefore, in the current study, emphasis has been given on integration of heuristic algorithms and random forest (RF) i.e., “RF-particle swarm optimization (PSO)”, “RF-grey wolf optimizer (GWO)” and “RF-grasshopper optimization algorithm (GOA)”, to identify groundwater vulnerable zones on the basis of field based hydrogeochemical parameters. In addition, correspondence health hazard of this area was assessed through human health hazard index. The spatial distribution of groundwater vulnerability revealed that middle-eastern and north-western part of the study area covered by very high and high, whereas central, western and south-western part are covered by very low and low vulnerability zones in outcomes of all the applied models. The evaluation result indicates that RF-GOA (AUC = 0.911) model performed the best considering testing dataset, and thereafter RF-GWO, RF-PSO and RF with AUC value is 0.901, 0.892 and 0.812 respectively. Findings also revealed the groundwater in this study region is quite unfavorable for drinking and irrigation purposes. The suggested models demonstrate their usefulness in foretelling sustainable groundwater resource management in various deltaic regions of the world through taking appropriate measures by policy-makers.

Traffic contributes to fine particulate matter (PM₂.₅) in the atmosphere through engine exhaust emissions and road dust generation. However, the evolution of traffic related PM₂.₅ emission over recent years remains unclear, especially when various efforts to reduce emission e.g., aftertreatment technologies and high emission standards from China IV to China V, have been implemented. In this study, hourly elemental carbon (EC), a marker of primary engine exhaust emissions, and trace element of calcium (Ca), a marker of road dust, were measured at a nearby highway sampling site in Shanghai from 2016 to 2019. A random forest-based machine learning algorithm was applied to decouple the influences of meteorological variables on the measured EC and Ca, revealing the deweathered trend in exhaust emissions and road dust. After meteorological normalization, we showed that non-exhaust emissions, i.e., road dust from traffic, increased their fractional contribution to PM₂.₅ over recent years. In particular, road dust was found to be more important, as revealed by the deweathered trend of Ca fraction in PM₂.₅, increasing at 6.1% year⁻¹, more than twice that of EC (2.9% year⁻¹). This study suggests that while various efforts have been successful in reducing vehicular exhaust emissions, road dust will not abate at a similar rate. The results of this study provide insights into the trend of traffic-related emissions over recent years based on high temporal resolution monitoring data, with important implications for policymaking.

Air pollution is becoming serious in developing country, and how to quantify the role of local emission and/or meteorological factors is very important for government to implement policy to control pollution. Here, we use a random forest model, a machine learning (ML) approach, combined with a de-weather method to analyze the PM₂.₅ level during the COVID-19 outbreak in Hubei Province. The results show that changes in anthropogenic emissions have reduced PM₂.₅ concentrations in February and March 2020 by about 33.3% compared to the same period in 2019, while changes in meteorological conditions have increased PM₂.₅ concentrations by about 8.8%. Moreover, the impact of meteorological conditions is more significant in the central region, which is likely to be related to regional transport. After excluding the contribution of meteorological conditions, the PM₂.₅ concentration in Hubei Province in February and March 2020 is lower than the secondary standard of China (35 μ g/m³). Our estimates also indicate that under similar meteorological conditions as in February and March 2019, an emission reduction intensity equivalent to about 48% of the emission reduction intensity during the lockdown may bring the annual average PM₂.₅ concentration to the standard (35 μ g/m³). Our study shows that machine learning is a powerful tool to quantify the influencing factors of PM₂.₅, and the results further emphasize the need for scientific emission reduction as well as joint regional control measures in future.

The characterisation of nanoplastics is much more difficult than that of microplastics. Herewith we employ Raman imaging to capture and visualise nanoplastics and microplastics, due to the increased signal-noise ratio from Raman spectrum matrix when compared with that from a single spectrum. The images mapping multiple characteristic peaks can be merged into one using logic-based algorithm, in order to cross-check these images and to further increase the signal-noise ratio. We demonstrate how to capture and identify microplastics, and then zoom down gradually to visualise nanoplastics, in order to avoid the shielding effect of the microplastics to shadow and obscure the nanoplastics. We also carefully compare the advantages and disadvantages of Raman imaging, while giving recommendations for improvement. We validate our approach to capture the microplastics and nanoplastics as particles released when we cut and assemble PVC pipes in our garden. We estimate that, during a cutting process of the PVC pipe, thousands of microplastics in the range of 0.1–5 mm can be released, along with millions of small microplastics in the range of 1–100 μm, and billions of nanoplastics in the range of <1 μm. Overall, Raman imaging can effectively capture microplastics and nanoplastics.

Air pollution and extreme heat have been responsible for more than a million deaths in China every year, especially in densely urbanized regions. While previous studies intensively evaluated air pollution episodes and extreme heat events, a limited number of studies comprehensively assessed atmospheric hot-and-polluted-episodes (HPE) – an episode with simultaneously high levels of air pollution and temperature – which have potential adverse synergic impacts on human health. This study focused on the Pearl River Delta (PRD) region of China due to its high temperature in summer and poor air quality throughout a year. We employed geostatistical downscaling to model meteorology at a spatial resolution of 1 km, and applied a machine learning algorithm (XGBoost) to estimate a high-resolution (1 km) daily concentration of particulate matter with an aerodynamic diameter ≤2.5 μm (PM₂.₅) and ozone (O₃) for June to October over 20 years (2000–2019). Our results indicate an increasing trend (∼50%) in the frequency of HPE occurrence in the first decade (2000–2010). Conversely, the annual frequency of HPE occurrence reduced (16.7%), but its intensity increased during the second decade (2010–2019). The northern cities in the PRD region had higher levels of PM₂.₅ and O₃ than their southern counterparts. During HPEs, regional daily PM₂.₅ exceeded the World Health Organization (WHO) and Chinese guideline levels by 75% and 25%, respectively, while the O₃ exceeded the WHO O₃ standard by up to 69%. Overall, 567,063 (95% confidence interval (CI): 510,357–623,770) and 52,231 (95%CI: 26,116–78,346) excessive deaths were respectively attributable to exposure to PM₂.₅ and O₃ in the PRD region. Our findings imply the necessity and urgency to formulate co-benefit policies to mitigate the region's air pollution and heat problems.

The San Joaquin Valley (SJV) of California has suffered persistent particulate matter (PM) pollution despite many years of control efforts. To further understand the chemical drivers of this problem and to support the development of State Implementation Plan for PM, a time-of-flight aerosol chemical speciation monitor (ToF-ACSM) outfitted with a PM₂.₅ lens and a capture vaporizer has been deployed at the Fresno-Garland air monitoring site of the California Air Resource Board (CARB) since Oct. 2018. The instrument measured non-refractory species in PM₂.₅ continuously at 10-min resolution. In this study, the data acquired from Oct. 2018 to May 2019 were analyzed to investigate the chemical characteristics, sources and atmospheric processes of PM₂.₅ in the SJV. Comparisons of the ToF-ACSM measurement with various co-located aerosol instruments show good agreements. The inter-comparisons indicated that PM₂.₅ in Fresno was dominated by submicron particles during the winter whereas refractory species accounted for a major fraction of PM₂.₅ mass during the autumn associated with elevated PM₁₀ loadings. A rolling window positive matrix factorization analysis was applied to the organic aerosol (OA) mass spectra using the Multilinear Engine (ME-2) algorithm. Three distinct OA sources were identified, including vehicle emissions, local and regional biomass burning, and formation of oxygenated species. There were significant seasonal variations in PM₂.₅ composition and sources. During the winter, residential wood burning and oxidation of nitrogen oxides were major contributors to the occurrence of haze episodes with PM₂.₅ dominated by biomass burning OA and nitrate. In autumn, agricultural activities and wildfires were found to be the main cause of PM pollution. PM₂.₅ concentrations decreased significantly after spring and were dominated by oxygenated OA during March to May. Our results highlight the importance of using seasonally dependent control strategies to mitigate PM pollution in the SJV.

Source apportionment can be an effective tool in mitigating soil pollution but its efficacy is often limited by a lack of information on the factors that influence the accumulation of pollutants at a site. In response to this limitation and focusing on a suite of heavy metals identified as priorities for pollution control, the study established a comprehensive pollution control framework using factor identification coupled with spatial agglomeration for agricultural soils in an industrialized part of Zhejiang Province, China. In addition to elucidating the key role of industrial and traffic activities on heavy metal accumulation through implementing a receptor model, specific influencing factors were identified using a random forest model. The distance from the soil sample location to the nearest likely industrial source was the most important factor in determining cadmium and copper concentrations, while distance to the nearest road was more important for lead and zinc pollution. Soil parent materials, pH, organic matter, and clay particle size were the key factors influencing accumulation of arsenic, chromium, and nickel. Spatial auto-correlation between levels of soil metal pollution and industrial agglomeration can enable a more targeted approach to pollution control measures. Overall, the approach and results provide a basis for improved accuracy in source apportionment, and thus improved soil pollution control, at the regional scale.